ConwayLife.com

So, you probably know the almost knightship. It acts like an elementary knightship akin to sir Robin or the sprayer, but it's back has a slight variation at generation 6. This causes it to disintegrate.
x = 13, y = 19, rule = B3/S23
11bo$6b2o2bo$6bob2o$5bo5b2o$bobo2bo$o$o3bo2b3ob2o$o3bo4bo ...

#01 *
#02 *
#03 *
#04 ***
#05 **
#06 *
#07 *
#08 ***
#09 ***
#10 ***
#11 **
#12 *
#13 *
#14 **
#15 *
#16 *
#17 **
#18 **
#19 ***
#20 **
#21 ***
#22 *
#23 *
#24 *
#25 **
#26 *
#27 **
#28 *
#29 *
#30 *
#31 **
#32 *
#33 ***
#34 **
#35 *
#36 ***
#37 **

I found a new c/5 orthogonal ship using ikpx2. Width 29 (27 in two generations).
I started out with this head partial:
x = 19, y = 31, rule = B3/S23
7b2ob2o$4b2obo3bob2o$4b2obobobob2o$7bo3bo$7bo3bo$3ob4obob4ob3o$o7bobo
7bo$9bo$o4bo7bo4bo$4o2bo5bo2b4o$2b2o2b2o3b2o2b2o$5b3o3b3o$2b2o2bo5bo2b
2o ...

The only w10 gutter c/7 ship is two loafers (and "two loafers but they're each flipped the other way"), or technically could also be loafer pushalongs that were not output due to qfind's duplicate row elimination. Attached the last dump (it's a dumpfile, not a .txt, but the forum didn't let me ...

Can you do echo $SHELL ?

It's bash.
$ echo $SHELL
/bin/bash


Then i don't know what the problem may be. It looks to be some kind of failed regex parse at the very beginning of the script. You can try to remove the first " rulearch= " line (5) and the following " if " and "fi " lines (7 and ...

You're right about the recompile.sh size. I still seem to have a version from an older branch i'm using...

I verified just now that the master branch of ikpx2 on gitlab compiles on a Cygwin installation on one of my older Win10 workstations.
I do get the grep: warning: stray \ before - message, so ...

Resu wrote: August 17th, 2025, 4:04 am Does anyone know how to fix this?

Your recompile.sh script has 64 lines? The script in the master branch has 47 lines. Have you done a recent pull, or did you modify any files?
And what system are you trying to build ikpx2 on?

Some time ago i have made a modification to the JLS source code to search for diagonal glide-symmetric ships. The git repository is at https://github.com/frankeverdij/JLSsource in the glide_code branch.

You can build this version yourself by obtaining a Java Development Kit: There is a compile.txt ...

I've known about this bug for a long time, and it only seems to occur when the search finishes at an extremely shallow depth with no spaceships found. I suspect this is a result of printing the pattern buffer before it's been written.

Correct. Valgrind confirms that patternBuf is used before ...

FYI: While testing the WLS smartsearch routine i ported in lifesrc to be able to run on the command-line, i received a false positive when searching for a 2c/5 spaceship in life rule. After a day of debugging the code and finding nothing, i ran WLS (Both K and J versions) and could confirm the false ...

This is a new orthogonal c/5 ship i discovered by continuing the earlier qfind -w13 gutter search. I found an idle memory node and figured i might do another 24 hour run. It popped up one hour after the start of the batch run, along with more copies of Kermit.
x = 25, y = 9, rule = B3/S23
4bo15bo ...

I think the most impressive speedups and resulting discoveries yet to happen will come from new algorithms (or new applications of algorithms previously known from other math/CS fields), rather than from just throwing enormous brute-force resources.

If an algorithm is scalable and expandable ...

The header file common.h is large and complicated enough that I should probably split it into multiple files, but I'm not sure what a good structure for this is. I'll start looking into it, but if anyone has a link to a good resource on this, please let me know.

I haven't come across an easy ...

If you don't get around to this, do you mind if I incorporate some of your changes anyway? Specifically, the #ifdef _OPENMP wrapping and backporting to c?

I saw your updates to the master branch and prepared two patch files yesterday, one for the _OPENMP wrapping (small) and one for the c ...

Please let me know if you have any comments or suggestions and especially if you find any bugs.


I'm currently reviewing your latest commits to my fork and am running some benchmark tests. I also have some quality-of-life improvements, like adding a Makefile and wrapping all thread/openmp ...

#01 ***
#02 **
#03 **
#05 *
#06 *
#07 **
#08 ***
#09 ***
#10 **
#11 **
#12 **
#13 **
#14 **
#15 **
#16 *
#17 *
#19 *
#21 **
#23 **
#24 *
#25 *
#26 **
#27 ***
#28 *
#30 **
#32 *
#33 ***
#34 *

Almost c/7d tubstretcher:

I noticed that all of your c/7d partials are symmetrical. Are you using ikpx2? If so, have you made any customisations to it?
In my ikpx2 searches of c/7d most of the partials which are logged are asymmetrical. This is fine, but i don't see a way to enforce symmetry if ...

139P7H1V0 can also support Taco partials:
x = 45, y = 37, rule = B3/S23
bo41bo$obo39bobo$o2bo13bo9bo13bo2bo$3o16bo5bo16b3o$b2o11bo5bo3bo5bo11b
2o$21bobo$2o11b2ob3ob2ob2ob3ob2o11b2o$2bo39bo$2o2b2o11bo9bo11b2o2b2o$
4b2o10bobo7bobo10b2o$3bobo9bo4b2ob2o4bo9bobo$2b3o8b2o3b4ob4o3b2o8b3o ...

I've completed another c/7 partial in the collection i posted earlier:
x = 19, y = 56, rule = B3/S23
6bo5bo$5bobo3bobo$4bo2bo3bo2bo$5b2o5b2o2$7bo3bo$5b4ob4o2$5bobo3bobo$8b
obo$5b2o5b2o$6bo5bo$7bo3bo$7bo3bo$7bo3bo$8b3o$4bo3b3o3bo$3b3o7b3o$2bo
13bo$3b3o7b3o$4b2o7b2o$4bo9bo$6bo5bo$bo5bo3bo5bo$8bobo ...

This is my current set of c/7 partials i amassed over 1.5 years:
#CXRLE Pos=0,0
x = 416, y = 261, rule = LifeHistory
7.A5.A17.D5.D13.A5.A16.A5.A21.A5.A14.A5.A14.A5.A12.A5.A15.A5.A16.A13.
A14.A9.A13.A9.A11.D9.D12.A9.A15.A2.A21.4A18.A10.A11.A9.A$6.A.A3.A.A
15.D.D3.D.D11.A.A3.A.A14.A.A3.A.A19.A.A3.A ...

The front spark is really cool and makes more pushalongs possible:
x = 26, y = 102, rule = B3/S23
23b3o$20bo$19bo4bo$20bo$21b2o$21bobo$21bob2o$19bob2o$19bobo2$21bobo$
21bobo$21b2o2$16b3o$16b3o$16bobo$15bo2b2o$14bo3bob2o$14bo4b3o$13bo2bo
4bo$16bo2$17b5o$14bo2b5o$15b2o2b4o$14b2o$15bo7bo$14bo7bo ...

Look what I found!

Well done! What program did you use?

The front spark is really cool and makes more pushalongs possible:
x = 26, y = 102, rule = B3/S23
23b3o$20bo$19bo4bo$20bo$21b2o$21bobo$21bob2o$19bob2o$19bobo2$21bobo$
21bobo$21b2o2$16b3o$16b3o$16bobo$15bo2b2o$14bo3bob2o$14bo4b3o$13bo2bo ...

C_R_116 wrote: April 15th, 2024, 1:14 am Known c/6d tagalong? (Check the top:)

I haven't seen this before, so it's not previously known. Nice find!

Why is it that whenever I try to find a width 11 wave supporting c/7, I never find a solution but I find an extremely large partial:


I can relate: See the c/7 partials over here .
Very typical how extensions of this speed are so easily found, but no tails whatsoever.

My hunch is that it is ...

... Is there anyone who thinks the entries should not be separated?

I think they should not be separated and that they belong together, because wickstretchers and greystretchers share the same principals. I don't think you need to differentiate/separate if a wick consists of one tub or a ...